Probabilistic Jacobian-based Saliency Maps Attacks

2 code implementations • 12 Jul 2020 • Théo Combey, António Loison, Maxime Faucher, Hatem Hajri

Neural network classifiers (NNCs) are known to be vulnerable to malicious adversarial perturbations of inputs including those modifying a small fraction of the input features named sparse or $L_0$ attacks.

Autonomous Driving Malware Detection